Introduction

A 12-year-old female patient diagnosed with double-chamber right ventricle and cerebral palsy was admitted for routine control. The patient’s history revealed that she had used antibiotics with a complaint of high fever 1 month ago and had intermittent fever and vomiting for the last few days. Physical examination revealed low-grade fever, cardiac murmur, and neuromuscular retardation but no haemodynamic disturbance. Previous transthoracic echocardiography was performed 4 months ago, and it showed severe subvalvular pulmonary stenosis (pulmonary valve velocity: 5 m/s) due to a double-chamber right ventricle. She also had severe degree tricuspid valve insufficiency, mitral valve prolapse with mild degree mitral regurgitation, and a tiny patent ductus arteriosus. At the last admission, we noticed that there were multiple echogenicities compatible with vegetations on the tricuspid (12 × 13 mm) and pulmonary valves (11 × 7 mm) (Figure 1, Video 1). These echocardiographic findings were accepted as major echocardiographic criteria of endocarditis. The acute phase reactants were elevated. The patient was hospitalised with the diagnosis of infective endocarditis. The wide spectrum of antibiotics (ceftriaxone, vancomycin, and amikacin) were started immediately after blood cultures were taken. Six blood cultures were taken from our patient during hospitalisation, and Enterococcus faecium growth was detected in only one, so it was accepted as a minor criterion. Congenital cardiac abnormalities, elevated acute phase reactants, and positive blood cultures were accepted as three minor criteria of endocarditis.

Figure 1. Transthoracic echocardiography showed vegetations in the tricuspid and pulmonary valves.

Although she has no haemodynamic disturbance in clinical evaluation, we observed the persistence of the vegetation with no change in size and the low grade of fever peaks in spite of 22 days of antibiotics at the follow-up. Then an open-heart surgery was performed. The vegetation on the tricuspid valve was removed with surgery, the tricuspid valve was repaired, the vegetation on the pulmonary valve was cleared, the cleft on the pulmonary valve was repaired, and the ductus was ligated. Muscle band excision was performed through the right ventricle side by deep myectomy. Postoperative transthoracic echocardiogram showed no vegetations on the valves, no obstruction of the right ventricular outflow tract (gradient 15 mmHg), and only mild tricuspid and pulmonary regurgitation. There was a small perimembranous ventricular septal defect by colour and continuous wave Doppler echocardiography that could not be seen in preoperative echocardiography.

Furthermore, there was three colour Doppler left to right jet crossing the interventricular septum resembling ventricular septal defect, but a continuous low velocity flow by continuous wave Doppler was detected on the right ventricle outflow tract on surgical resection site of the muscle bound that was consistent with multiple coronary fistulas (Figure 2, Video 2). Our patient had no signs of volume overload, and there were no findings related to coronary ischaemia by laboratory investigation (normal electrocardiogram and normal troponin level). After 1 month, echocardiography showed only one small size coronary fistula on the interventricular septum and spontaneous closure of the other two fistulas (Figure 2, Video 2). After 4 months, she has no medication and is free of symptoms. At the fourth month follow-up, the last coronary fistula disappeared spontaneously (Video 2).

Figure 2. Echocardiography showed three coronary fistulas in the early postoperative period and the single remaining coronary fistula at the first month postoperative visit.